Hyponatremia is a common electrolyte disorder observed in clinical settings [1,2,3,4], particularly in critically ill and older patients. It is associated with poor clinical outcomes such as increased mortality, disability, prolonged hospital stay, and increased hospital costs [16,17,18]. Published data show that even mild and chronic hyponatraemia represents an economic burden [17, 18].
Previous studies on older patients have reported a prevalence of chronic hyponatremia of 8% among those living in private homes . Tay et al. measured the electrolytes of 5873 outpatients (age ≥ 60 years) and found that 403 (6.9%) had hyponatremia in at least one blood test . In a study of 118 residents (mean age 85.5 years) in a single geriatric health service facility, Naka et al. reported that 33 patients (28%) had hyponatremia . In a retrospective study of inpatients at a tertiary hospital, Amit et al. reported that the incidence of hyponatremia in older patients (average age 73.9 ± 6.5 years) was 26.0% . In a study of 8377 emergency patients, Lmai et al. reported a significantly higher prevalence of hyponatremia among older patients (≥ 65 years old, 17.0%) than among younger adults (18–64 years, 5.7%). Moreover, in that study, the prevalence of severe hyponatremia (serum sodium < 125 mmol/L) was significantly higher among older patients relative to younger adults (1.9% vs. 0.3%) . In a comparative study of 2849 older patients (≥ 60 years) and 6756 non-older patients (13–59 years) at a single center during the same period, Siregar reported that the incidence of hyponatremia in older inpatients in a general ward setting was 2.43 times that observed in non-older patients (26.32% vs. 10.85%), while the incidence of HAH in older inpatients was 2.79 times that reported in non-older patients (7.35% vs. 2.64%) .
In our study, the overall prevalence of hyponatremia among older hospitalized patients admitted to a single center within a 4-year period was 24.7%. Most cases involved mild hyponatremia, in contrast to similar previous studies. This difference may be attributable to differences in the criteria used to define hyponatremia and the population selection protocol, which fully explained the high incidence of hyponatremia in older hospitalized patients. Among the 4364 hospitalized patients with hyponatremia included in this study, those with moderate and severe hyponatremia had a significantly higher age than those with mild hyponatremia, indicating that increasing age is associated with a gradual degeneration of the tissues and organs of the body, impairment of the mechanisms that maintain normal body fluid homeostasis, and increased vulnerability to hyponatremia-related diseases and drugs. Accordingly, severe hyponatremia was more likely to occur with age.
In this study, 2097 patients (48.1%) had CAH on admission and 2267 patients (51.9%) had HAH after admission. The proportion of CAH increased with the severity of hyponatraemia. As previously reported, severe hyponatremia was more frequently community-acquired than hospital-acquired [2, 25]. We speculate that this may be related to more active serum sodium monitoring and related treatment after admission.
Many diseases have been associated with the onset of hyponatremia. It has been reported that pneumonia, heart failure, liver cirrhosis, malignant tumors, stroke, and other diseases are all prone to hyponatremia [26,27,28,29,30]. In our study, we analyzed the incidence of primary diseases in 4364 older patients with hyponatremia and identified respiratory diseases, tumors, cardiovascular diseases, central nervous system diseases, and orthopedic diseases as the most common, among which pulmonary infection, lung cancer, heart failure, cerebral infarction, and fracture accounted for the highest proportion. Most of the patients in our study had clinical euvolemia, which was likely partially attributable to the large number of patients (n = 1240) who met criteria for SIADH. In our study, SIADH was the most common cause of hyponatremia, accounting for 28.4%, which was similar to 31% reported by Amit et al. . Further research showed that the proportion of respiratory diseases is the highest in those with SIADH, followed by malignant tumors and central nervous system diseases. Our results for SIADH were similar to a previous study . Our findings indicate that the common causes of hyponatremia are respiratory system disease, tumor, cardiovascular system disease, central nervous system disease, etc. These diseases should be monitored carefully in older patients, and blood sodium concentrations should be monitored closely in clinical settings to ensure the earliest possible detection and correction of hyponatremia.
Many drugs used in clinical settings can also aggravate existing hyponatremia or induce iatrogenic hyponatremia. Several studies have reported that hyponatremia is caused by iatrogenic factors in 40–75% of patients . The mechanisms associated with drug-induced hyponatremia include volume clearance, the effects of sodium ion reabsorption in renal tubules and the stimulation of abnormal antidiuretic hormone (ADH) secretion or enhancement of its activity. Consistent with our findings, previous studies have identified several drugs associated with the development of hyponatremia, including diuretics (especially thiazide diuretics), dehydrants, PPIs, ACEIs/ARBs, SSRIs, carbamazepine and NSAIDs . In one report, 73%, 20% and 8% of cases of diuretic-related hyponatremia were induced by thiazide diuretics alone, a combination of thiazide and potassium-preserving diuretics or furosemide, respectively . Another study reported that 13.7% of 951 thiazide-treated outpatients had hyponatremia and that age > 70 years was associated with a 3.9 times increase in hyponatremia risk . Concomitant administration of drugs that affect water homeostasis, such as SSRIs, SNRIs, NSAIDs, or benzodiazepines, is frequent in older patients with thiazide-associated hyponatremia . In a retrospective study of 2569 patients (aged > 65 years) in community-dwelling older people, Grattagliano et al. reported the likelihood of hyponatremia trebled when four drugs were taken, and it was seven-fold higher with the use of six drugs . As mentioned earlier, our study identified PPIs, loop diuretics, potassium sparing diuretics, ACEIs/ARBs, thiazide diuretics and NSAIDs as drugs frequently associated with hyponatremia in patients. We further demonstrated that increasing hyponatremia severity was significantly associated with a higher frequency of loop diuretic and thiazide diuretic use, suggesting that these drugs are an important cause of hyponatremia in older patients. Moreover, we determined that the use of potassium-preserving diuretics and PPIs increased the incidence of moderate to severe hyponatremia and that the use of mannitol, ACEIs/ARBs, and NSAIDs increased the incidence of severe hyponatremia. These findings suggest the serum sodium concentrations of patients using these drugs in clinical settings should be monitored continuously, and the sodium intake should be increased appropriately as needed.
Early studies confirmed that hyponatremia is the most likely cause of deterioration and increased mortality associated with various internal environmental disorders [12, 13, 35, 36]. In a study of 53,236 hospitalized patients, Wald et al. noted reduced survival rates and prolonged hospitalization times among patients with CAH and HAH . Particularly, in patients with serum sodium concentrations < 138 mmol/L, the mortality risks associated with CAH and HAH increased by 8% and 23%, respectively, with each 1-mmol/L decrease in the serum sodium concentration. Tzoulis et al. found that patients with hyponatremia and a serum sodium concentration < 128 mmol/L had a mortality rate of 17.3% and a risk of death during hospitalization 3.3 times higher than that of patients with a serum sodium concentration > 128 mmol/L . Those findings suggest that hyponatremia is both an independent predictor of mortality and a direct cause of increased mortality. Consistent with those earlier findings, our analysis of older inpatients with different degrees of hyponatremia revealed that increasing severity was associated significantly with increases in the hospitalization time and mortality. This conclusion was reached by other scholars [1, 31]; however, it is at odds with the findings of Holland-Bill et al.. In that study, the risk of mortality increased with the decrease in serum sodium, but the decrease in serum sodium below a threshold of 132 mmol/L did not contribute to a further increase in overall mortality risk. Further, our results are in disagreement with the findings of the study by Zheng et al.. In this study, the in-hospital mortality was 1.7%, 11.1%, and 6.8% in the mild (≥ 130–< 135 mmol/L), moderate (≥ 125–< 130 mmol/L), and severe hyponatremia groups (< 125 mmol/L), respectively. Patients with moderate or severe hyponatremia had a higher mortality than those with mild hyponatremia (moderate vs. mild group: OR 6.92, 95% CI 2.53–18.92, P < 0.001; severe vs. mild group: OR 4.54, 95% CI 1.05–19.58, P = 0.043). There was no significant difference in the risk of death between the severe and moderate groups. In addition, in the specific analysis of primary diseases, the in-hospital mortality of patients with hyponatremia was higher in those with lung cancer, cerebral infarction and pulmonary infection (20.5%, 16.3%, 15.0%, respectively). The in-hospital mortality of heart failure patients was 11.7%, which was slightly lower than the above three diseases. We speculate that this may be related to the severity of primary diseases and the number of complications.
This study has several strengths. We studied a large and unselected population. As far as we know, this is the first study on the prevalence of hyponatremia in older inpatients in China. Our large research population enabled us to examine mortality risks associated with different levels of hyponatremia. However, our study has certain limitations. First, as this was a single center study with a relatively limited sample size, it was difficult to make precise clinico-etiological profiles and their association with different characteristics. Second, there was no data on post discharge mortality; this restricted the evaluation of in-hospital mortality, and therefore, we could not follow up on patients to assess long-term impact. Third, we did not have information regarding the use of outpatient drugs that were often associated with hyponatremia. Data including these drugs would enhance the information provided in our study. Fourth, The single serum sodium measurement in our study did not reflect changes in clinical management.